What makes a good pollinator?

This chapter discusses pollinator efficiency; pollination performance (from the perspective of the bee); pollinator dependency (from the perspective of the plant), including topics on breeding systems, as well as flower and fruit morphology; and pollinator performance (from the perspective of foraging ecology), including the taxon-based differences in bee flight distance, morphological considerations, and forager behaviour in rich and poor habitats.

Pollen tube growth from multiple pollinator visits more accurately quantifies pollinator performance and plant reproduction

Pollination services from animals are critical for both crop production and reproduction in wild plant species. Accurately measuring the relative contributions of different animal taxa to pollination service delivery is essential for identifying key pollinators. However, widely used measures of pollinator effectiveness (e.g., single visit pollen deposition) may be inaccurate where plant reproduction is strongly constrained by pollen quality. Here, we test the efficacy of single and multiple pollinator visits for measuring pollinator performance in a model plant species (apple, Malus domestica Borkh) that is strongly limited by pollen quality. We determined pollination success using a suite of measures (pollen deposition, pollen tube growth, fruit and seed set) from single and multiple pollinator visits. We found that pollen deposition from a single pollinator visit seldom resulted in the growth of pollen tubes capable of eliciting ovule fertilisation and never resulted in fruit or seed production. In contrast, multiple pollinator visits frequently initiated the growth of pollen tubes capable of ovule fertilisation and often led to fruit and seed production. Our findings suggest that single visit pollen deposition may provide a poor measure of pollinator performance when linked to reproductive success of plant species that are constrain by pollen quality. Alternatively, pollen tube growth from single and multiple pollinator visits can provide a measure of pollinator performance that is more closely linked to plant reproduction.

Effects of ants on pollinator performance in a distylous pericarpial nectary-bearing Rubiaceae in Brazilian Cerrado

Besides the eff ectiveness of floral visitors, to better understand pollination systems is necessary to consider the role of predators. Ants are ubiquitous on the vegetation, especially on plants bearing extrafloral (EFNs) and pericarpial nectaries (PNs). Both EFNs and PNs reward ants which in turn provide to plants effective protection against herbivores. However, ants can also repel pollinators and cause an indirect cost for the plant partner, although the role of ants on pollinators’ performance has rarely been assessed in Neotropics, mainly on PN-bearing plants. Here, our main aim was, through an experimental field study in terms of ant’s presence versus absence, to test the hypothesis that ants dissuade floral visitors by decreasing the time spent during visits on the PN-bearing Declieuxia fruticosa. Additionally, we recorded floral phenology, and quantified and qualified floral visits. We showed that bees were the most frequent pollinators and the presence of ants dissuades them. In ant presence, pollinators were on average 30% faster than without ants. Since D. fruticosa produces fruits mainly after cross-pollination, the role of ants may be profi table to plants as they induce pollinators to do shorter visits and search for other fl owers in conspecifi c plants. Therefore, pollinators avoid stay at longer on plants with ants in order to avoid attacks, which may contribute to plant outcrossing. However, whether positive or negative the effects of ants on D. fruticosa reproduction are, they remain to be studied.

Estimating pollinator performance of visitors to the self-incompatible crop-plant Brassica rapa by single visit deposition and pollen germination: a comparison of methods.

Estimating the pollen-deposition effectiveness of flower visitors is fundamental to understanding their performance as pollinators. While estimates of visitation rates, pollen loads, and single visit deposition (SVD) are all useful proxies for performance, and so help to reveal the relative effectiveness of different visitors, none take into account the breeding system of the plants, or the quality of pollen deposited. Here we compare the performance of visitors to the self-incompatible plant Brassica rapa (turnip) using SVD and pollen germination. We also report the first use of the staining of Brassica rapa stigma papilla cells (known to reveal a specific reaction to self-pollen) to compare self-pollen deposition between insect visitors. We found that most of the pollen grains deposited by insect visitors (and therefore counted by SVD methods) were non-germinating self-pollen. A smaller proportion of grains were outcrossed and so germinated. There was also a significant positive relationship between environmental conditions (wind speed) and pollen deposition, but not pollen germination.Both methods identified Bombus spp. as the best-performing visitors on turnip flowers, followed by Eristalis spp., whereas performance estimates for Episyrphus balteatus and ‘other hoverflies’ were no higher than controls for both methods. This study provides further insight into the methodology for estimating pollinator performance, especially in plants when only cross-pollen can germinate.

Pollinator importance networks illustrate the crucial value of bees in a highly speciose plant community

Accurate predictions of pollination service delivery require a comprehensive understanding of the interactions between plants and flower visitors. To improve measurements of pollinator performance underlying such predictions, we surveyed visitation frequency, pollinator effectiveness (pollen deposition ability) and pollinator importance (the product of visitation frequency and effectiveness) of flower visitors in a diverse Mediterranean flower meadow. With these data we constructed the largest pollinator importance network to date and compared it with the corresponding visitation network to estimate the specialisation of the community with greater precision. Visitation frequencies at the community level were positively correlated with the amount of pollen deposited during individual visits, though rarely correlated at lower taxonomic resolution. Bees had the highest levels of pollinator effectiveness, with Apis, Andrena, Lasioglossum and Osmiini bees being the most effective visitors to a number of plant species. Bomblyiid flies were the most effective non-bee flower visitors. Predictions of community specialisation (H2′) were higher in the pollinator importance network than the visitation network, mirroring previous studies. Our results increase confidence in existing measures of pollinator redundancy at the community level using visitation data, while also providing detailed information on interaction quality at the plant species level.